A method for verifying osteotomy accuracy in knee replacement surgery

Abstract

The present invention discloses a method for verifying osteotomy accuracy during knee replacement surgery, comprising: S1, obtaining a preoperative point cloud of a patient and marking anatomical bony landmarks therein; S2, determining whether it is a non-imaging surgical plan; if so, obtaining a planned osteotomy plane and a planned osteotomy amount, and simultaneously registering the patient's preoperative point cloud with the intraoperative point cloud; otherwise, obtaining a planned osteotomy plane and a planned osteotomy amount; S3, obtaining a postoperative point cloud of the patient after osteotomy, identifying the actual osteotomy plane therein, and then calculating the actual osteotomy amount, while removing points on the identified actual osteotomy plane; S4, registering the patient's preoperative point cloud with the postoperative point cloud after removing the points on the actual osteotomy plane; S5, based on the registration in S4, transforming the actual osteotomy plane obtained in S3 and the planned osteotomy plane obtained in S2 to the same reference, and calculating the osteotomy plane error and the osteotomy amount error accordingly. The present invention can verify osteotomy accuracy under optical navigation.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method for verifying osteotomy accuracy in knee replacement surgery. Background Art

[0002] Knee replacement surgery involves resection of the diseased distal femoral condyles (medial and posterior femoral condyles), as well as the articular surface of the tibial plateau, and replacement with an artificial prosthesis. Precise osteotomy and prosthesis selection during knee replacement are crucial for re-aligning the lower limb and restoring postoperative function. Malalignment is a major cause of prosthesis failure and revision after knee replacement. Therefore, precise osteotomy thickness and angle, as well as appropriate prosthesis size and placement, are crucial for a successful knee replacement.

[0003] In order to ensure that the patient's knee joint can maintain good motor function after replacement, it is necessary to introduce a method to evaluate the effect of knee replacement surgery. Summary of the Invention

[0004] Purpose of the invention: In view of the above-mentioned shortcomings, the present invention provides a method for verifying the osteotomy accuracy of knee replacement surgery, which can realize the evaluation of osteotomy accuracy under optical navigation, and ensure that the patient's knee joint can maintain relatively good motor function after replacement.

[0005] Technical solution:

[0006] The present invention provides a method for verifying osteotomy accuracy in knee replacement surgery, comprising:

[0007] S1. Obtain the patient's preoperative femoral and tibia model point cloud and mark the anatomical bony landmarks in it;

[0008] S2, determine whether it is a non-imaging surgery plan;

[0009] If yes, obtain the patient's intraoperative femoral and tibia model point clouds and the anatomical bony landmarks therein, and use them to obtain the planned osteotomy plane and planned osteotomy amount. At the same time, align the patient's preoperative femoral and tibia model point clouds with the intraoperative femoral and tibia model point clouds;

[0010] Otherwise, the planned osteotomy plane and planned osteotomy amount are obtained according to S1;

[0011] S3, obtaining the postoperative femoral and tibia model point cloud of the patient after osteotomy, identifying the actual osteotomy plane therein, and then calculating the actual osteotomy amount, while removing the points on the identified actual osteotomy plane;

[0012] S4, registering the patient's preoperative femoral and tibia model point clouds with the postoperative femoral and tibia model point clouds after removing points on the actual osteotomy plane;

[0013] S5. Based on the registration in S4, the actual osteotomy plane obtained in S3 and the planned osteotomy plane obtained in S2 are transformed to the same reference, and the osteotomy plane error and the osteotomy amount error are calculated accordingly.

[0014] Specifically, in S3, the actual osteotomy plane is identified by acquiring normal information and topological structure of points in the point cloud of the patient's postoperative femur and tibia models.

[0015] More specifically, in S3, identifying the actual osteotomy plane specifically includes:

[0016] S31, using a voxel block of a set size to perform voxel downsampling on each point in the postoperative femoral and tibia model point cloud of the patient, traversing each sampling point obtained after the voxel downsampling, and classifying a sampling point and points within a set range thereof as the same region of interest;

[0017] S32, clustering the region of interest obtained in S31 to obtain several candidate regions, and fitting several candidate planes based on the regions;

[0018] S33, screening a set of planes that meet the requirements from the candidate planes obtained in S32 as the final actual osteotomy plane.

[0019] Furthermore, in S32, clustering the regions of interest includes:

[0020] (1) Obtain the points and their normal vectors in each region of interest, fit each region of interest to a plane, and retain the regions of interest whose corresponding planes meet the set conditions;

[0021] (2) Clustering the regions of interest retained in step (1), merging them to obtain several candidate regions, and fitting each candidate region to obtain several candidate planes.

[0022] Furthermore, in step (1), the points and their normal vectors in each region of interest are obtained, and each region of interest is fitted into a plane using the least squares method by calculating the sum of the normal vectors of all points.

[0023] Furthermore, in step (1), the setting conditions are specifically: the distance between all points in each region of interest and the corresponding plane is less than a first set distance, and the angle between the sum of the normal vectors of all points in each region of interest and the normal vector of the corresponding plane is less than a first set angle.

[0024] Furthermore, in step (2), the clustering conditions include:

[0025] For any two regions of interest A and B, the distance between the points in A and the points in B is traversed and calculated. When the distance between the nearest points in A and B, that is, the aforementioned minimum distance, is less than the second set distance, and the angle between the normal vectors of the planes corresponding to A and B is less than the second set angle, and at the same time, the average distance between all points in A and the plane corresponding to B is less than the third set distance, then the regions of interest A and B are considered to be the same category, and the regions of interest A and B are merged into a new region as a candidate region.

[0026] Furthermore, in S33, the screening requirement is that the number of points corresponding to the candidate plane is greater than a set number, and the area of ​​the minimum circumscribed rectangle of all points corresponding to the candidate plane is greater than a set area.

[0027] Specifically, in S4, the point cloud of the patient's preoperative femur and tibia model is defined as the preoperative point cloud, and the point cloud of the patient's postoperative femur and tibia model after removing the points on the actual osteotomy plane is defined as the postoperative point cloud;

[0028] Then the S4 includes:

[0029] S41. Randomly select at least three point pairs at corresponding positions of the pre-operative point cloud and the post-operative point cloud, and calculate the initial transformation matrix of the two.

[0030] S42. Using the initial transformation matrix obtained in S41 as an initial value, use the ICP algorithm to calculate the transformation matrix between the pre-operative point cloud and the post-operative point cloud.

[0031] More specifically, in S41, after obtaining the initial transformation matrix, the step of optimizing the initial transformation matrix is ​​further included, including:

[0032] 1) transforming the postoperative point cloud according to the initial transformation matrix to the same reference as the preoperative point cloud;

[0033] 2) Randomly select three points in the preoperative point cloud whose distance between each pair is greater than a set value, and find the point in the transformed postoperative point cloud that is closest to the selected point, thereby obtaining three sets of point pairs;

[0034] 3) Repeat step 2) a set number of times, selecting three point pairs each time for registration, calculating the corresponding transformation matrix, and applying it to the postoperative point cloud;

[0035] 4) Calculate the average distance between the three points selected in the preoperative point cloud and the corresponding points in the postoperative point cloud transformed in step 3), and select the transformation matrix corresponding to the minimum average distance as the initial transformation matrix between the final preoperative point cloud and the postoperative point cloud.

[0036] Specifically, in S5, the osteotomy accuracy is verified by calculating the osteotomy plane errors and osteotomy amount errors of the distal femoral osteotomy plane, the posterior femoral condyle osteotomy plane, and the tibial plateau osteotomy plane.

[0037] More specifically, the calculation of the osteotomy plane error and the osteotomy amount error of the distal femoral osteotomy plane, the posterior femoral condyle osteotomy plane, and the tibial plateau osteotomy plane includes:

[0038] S51. Based on the registration in S4, transform the actual osteotomy planes in the point clouds of the postoperative femoral and tibia models of the patient, transform the actual osteotomy planes to the same reference as the planned osteotomy planes, and search for planes corresponding to the distal femoral osteotomy plane, the posterior femoral condyle osteotomy plane, and the tibial plateau osteotomy plane in the planned osteotomy planes.

[0039] S52, calculating the angle error and distance error between the actual osteotomy plane and the planned osteotomy plane according to S51;

[0040] S53: Combine S3 and S51 to obtain the actual osteotomy amount of the corresponding plane, and combine it with the planned osteotomy amount obtained in S2 to calculate the osteotomy amount error.

[0041] Furthermore, in the S52, the angle error and distance error between the actual osteotomy plane and the planned osteotomy plane are calculated as follows:

[0042] Calculate the average distance between all intersection points of the distal femoral osteotomy plane in the planned osteotomy plane and the patient's postoperative femoral and tibia model point cloud and the corresponding plane in the actual osteotomy plane, and calculate the angle between the normal direction of the distal femoral osteotomy plane in the planned osteotomy plane and the normal direction of the corresponding plane in the actual osteotomy plane;

[0043] Calculate the average distance between all intersection points of the femoral posterior condyle osteotomy plane in the planned osteotomy plane and the patient's postoperative femoral and tibial model point cloud and the corresponding plane in the actual osteotomy plane, and calculate the angle between the normal direction of the femoral posterior condyle osteotomy plane in the planned osteotomy plane and the normal direction of the corresponding plane in the actual osteotomy plane;

[0044] Calculate the average distance between all intersection points of the tibial plateau osteotomy plane in the planned osteotomy plane and the patient's postoperative femoral and tibia model point cloud and the corresponding plane in the actual osteotomy plane, and calculate the angle between the normal direction of the tibial plateau osteotomy plane in the planned osteotomy plane and the normal direction of the corresponding plane in the actual osteotomy plane;

[0045] The above three sets of distances and angles together describe the angle error and distance error between the actual osteotomy plane and the planned osteotomy plane.

[0046] Furthermore, in S53, the actual amount of osteotomy of the corresponding plane is obtained by combining S3 and S51, specifically:

[0047] Calculate the distances from the distal femoral medial condyle point and the distal femoral lateral condyle point to the distal femoral osteotomy plane in the planned osteotomy plane to obtain the distal femoral medial condyle osteotomy amount and the distal femoral lateral condyle osteotomy amount;

[0048] Calculate the distances from the medial point of the posterior femoral condyle and the lateral point of the posterior femoral condyle to the posterior femoral condyle osteotomy plane in the planned osteotomy plane to obtain the medial and lateral osteotomy amounts of the posterior femoral condyle;

[0049] The distances from the medial concave point of the tibial plateau and the lateral concave point of the tibial plateau to the tibial plateau osteotomy plane in the planned osteotomy plane were calculated to obtain the medial tibial plateau osteotomy amount and the lateral tibial plateau osteotomy amount.

[0050] Specifically, in S2, the registration of the patient's preoperative femur and tibia model point clouds with the intraoperative femur and tibia model point clouds is performed as follows:

[0051] The intraoperative anatomical bony landmarks and the preoperative anatomical bony landmarks obtained by S1 form corresponding point pairs, and the initial transformation matrix between the patient's preoperative femur and tibia model point cloud and the intraoperative femur and tibia model point cloud is calculated. Based on this, the ICP algorithm is used to align the patient's preoperative femur and tibia model point cloud with the intraoperative femur and tibia model point cloud to obtain the transformation matrix between the two.

[0052] Beneficial effects: The present invention can achieve registration with the point cloud of the defective bone model obtained after surgery, ensure the accuracy of the registration, and verify the accuracy of osteotomy under optical navigation. Regardless of whether there is an imaging surgery plan or a non-imaging surgery plan, the present invention can achieve accuracy evaluation to ensure that the patient's knee joint can maintain relatively good motor function after replacement. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0054] Figure 1 A flowchart of the osteotomy accuracy verification method for knee replacement surgery of the present invention;

[0055] Figure 2 This is an example diagram of the osteotomy plane of the present invention. DETAILED DESCRIPTION

[0056] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0057] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should have the common meanings understood by persons having ordinary skills in the field to which the present invention belongs.

[0058] Generally, before a knee replacement surgery, the doctor will make a surgical plan based on the patient's CT images, mark anatomical bony landmarks on the patient's femur and tibia, and determine the osteotomy plane based on these landmarks. During the operation, osteotomy will be performed along the osteotomy plane. This plan is an imaging surgery plan; while the non-imaging surgery plan is to directly mark the anatomical bony landmarks of the femur and tibia under the optical system to determine the osteotomy plane. Regardless of whether it is an imaging technology plan or a non-imaging technology plan, the basic principle is to generate a planned osteotomy plane based on the geometric relationship of the anatomical bony landmarks of the femur and tibia. Therefore, the present invention uses the error between the actual osteotomy amount and the planned osteotomy amount, as well as the plane angle and distance error between the planned osteotomy plane and the actual osteotomy plane to measure the osteotomy accuracy of knee replacement surgery.

[0059] The flowchart of the osteotomy accuracy verification method for knee replacement surgery of the present invention is as follows: Figure 1 Shown, including:

[0060] S1. Obtain the patient's preoperative images, perform segmentation and reconstruction on the patient's preoperative images, obtain the patient's preoperative femur and tibia model point clouds, and mark the anatomical bony landmarks on the femur and tibia;

[0061] In the present invention, the patient's preoperative femur and tibia model point clouds are complete femur and tibia model point clouds.

[0062] In the present invention, the anatomical bony landmarks include the center of the femoral head, the center of the femoral knee joint, the climbing point of the femoral shaft, the medial epicondyle of the femur, the lateral epicondyle of the femur, the distal lateral condyle of the femur, the distal medial condyle of the femur, the posterior lateral condyle of the femur, the posterior medial condyle of the femur, the medial concave point of the tibial plateau, the lateral concave point of the tibial plateau, the center of the tibial knee joint and the inner 1 / 3 point of the tibial tuberosity. The corresponding osteotomy plane can be obtained according to the geometric relationship of these landmarks.

[0063] S2, determine whether it is a non-imaging surgery plan;

[0064] If yes, obtain the intraoperative femur and tibia model point cloud of the patient and the anatomical bony landmarks on the femur and tibia, generate a planned osteotomy plane based on the geometric relationship of the anatomical bony landmarks, and calculate the planned osteotomy amount;

[0065] Based on the intraoperative anatomical bony landmarks and the preoperative anatomical bony landmarks obtained by S1, the patient's preoperative femur and tibia model point clouds are registered with the patient's intraoperative femur and tibia model point clouds, and the planned osteotomy plane is transformed into the patient's preoperative femur and tibia model point clouds;

[0066] Otherwise, the planned osteotomy plane is generated based on the anatomical bony landmarks on the femur and tibia marked in the point cloud of the patient's preoperative femur and tibia model obtained by S1, and the planned osteotomy amount is calculated at the same time;

[0067] In the present invention, the patient's intraoperative femur and tibia model point clouds are obtained during the operation, that is, the bone surface point clouds collected during the patient's operation are loaded; the anatomical bone landmark points on the femur and tibia of the patient's intraoperative femur and tibia model point clouds are marked and loaded.

[0068] In the present invention, the patient's preoperative femur and tibia model point cloud is aligned with the patient's intraoperative femur and tibia model point cloud. Specifically, corresponding point pairs are formed with the intraoperative anatomical bony landmark points and the preoperative anatomical bony landmark points obtained by S1, and the initial transformation matrix between the patient's preoperative femur and tibia model point cloud and the patient's intraoperative femur and tibia model point cloud is calculated. Based on this, the ICP algorithm is used to align the patient's preoperative femur and tibia model point cloud with the patient's intraoperative femur and tibia model point cloud to obtain the transformation matrix between the two, thereby completing the preoperative and intraoperative alignment.

[0069] In the present invention, in order to improve the calculation speed, before aligning the patient's preoperative femur and tibia model point cloud with the patient's intraoperative femur and tibia model point cloud, a step of voxel down sampling of the patient's intraoperative femur and tibia model point cloud is also included. In this embodiment, voxel down sampling can be used.

[0070] S3. Acquire an image of the patient after osteotomy, segment and reconstruct the image, obtain a point cloud of the femur and tibia model of the patient after osteotomy, identify the actual osteotomy plane, calculate the actual osteotomy amount, and remove points on the actual osteotomy plane;

[0071] In the present invention, the image of the patient after osteotomy is segmented and reconstructed to obtain a triangular patch model composed of a set of bone model point clouds, namely, the postoperative bone model point cloud and its normal vector, which is an incomplete femur and tibia model point cloud. When the patient's complete femur and tibia model point cloud before surgery is subsequently aligned with the incomplete femur and tibia model point cloud after surgery, the osteotomy plane identified in the patient's postoperative femur and tibia model point cloud may have no corresponding point in the patient's preoperative femur and tibia model point cloud. Therefore, before alignment, the actual osteotomy plane obtained in the postoperative image must be identified, and then the actual osteotomy amount must be calculated, and the points on the actual osteotomy plane must be removed.

[0072] In this embodiment, when removing points on the actual osteotomy plane, points within a set area with the actual osteotomy plane as the reference can be removed. The set area is an area with a distance less than 1.5-2 mm from the boundary of the actual osteotomy plane. In this embodiment, 2 mm is preferred. By such a design, the registration accuracy can be further ensured.

[0073] In the present invention, by obtaining the normal information and the topological structure of the midpoints in the point cloud of the patient's postoperative femoral and tibia models, the osteotomy plane can be identified, that is, the actual osteotomy plane can be obtained, as follows:

[0074] S31, using a voxel block of a set size to perform voxel downsampling on each point in the postoperative bone model point cloud, traversing each sampling point obtained after the downsampling, and classifying a sampling point and points within a set range thereof as the same region of interest;

[0075] In this embodiment, according to statistics, the minimum width of the osteotomy plane to be calculated is about 20 mm. In order to ensure that each surface participating in the calculation can be identified, the voxel block size is set to 10 mm×10 mm×10 mm.

[0076] In this embodiment, the setting range of a certain sampling point can specifically be a spherical range with the sampling point as the center and a radius of 5 mm.

[0077] S32, clustering the region of interest obtained in S31 to obtain several candidate regions, and fitting several candidate planes based on the regions;

[0078] In the present invention, the regions of interest obtained in S31 are clustered as follows:

[0079] (1) Obtaining the points and normal vectors within each region of interest obtained in S31, and then fitting each region of interest to obtain a plane, and retaining the regions of interest whose corresponding planes meet the set conditions;

[0080] Specifically, the points and normal vectors within each region of interest obtained in S31 are obtained, and the sum of the normal vectors of all points is calculated, thereby fitting each region of interest into a plane using the least squares method.

[0081] In the present invention, the setting conditions are specifically: the distance between all points in each region of interest and the corresponding plane is less than a first set distance, and the angle between the sum of the normal vectors of all points in each region of interest and the normal vector of the corresponding plane is less than a first set angle.

[0082] In this embodiment, the first set distance may be 1 mm, and the first set angle may be 1°.

[0083] (2) Clustering the regions of interest retained in step (1), merging them to obtain several candidate regions, and fitting each candidate region to obtain several candidate planes;

[0084] In the present invention, the conditions for clustering include:

[0085] For any two regions of interest A and B, the distances between points in A and points in B are traversed and calculated. When the distance between the nearest points in A and B, i.e., the aforementioned minimum distance, is less than the second set distance, and the angle between the normal vectors of the planes corresponding to A and B is less than the second set angle, and at the same time, the average distance between all points in A and the plane corresponding to B is less than the third set distance, then the regions of interest A and B are considered to be of the same category, and the regions of interest A and B can be merged into a new region as a candidate region.

[0086] In this embodiment, the second set distance may be 1 mm, and the second set angle may be 1°.

[0087] In this embodiment, the third set distance may be 2 mm.

[0088] In the present invention, the least square method is used to fit each candidate region.

[0089] S33, screening a set of planes that meet the requirements from the candidate planes obtained in S32 as the final actual osteotomy plane.

[0090] In the present invention, the screening requirements are: the number of points corresponding to the candidate plane is greater than a set number, and the area of ​​the minimum circumscribed rectangle of all points corresponding to the candidate plane is greater than a set area.

[0091] In this embodiment, the set number can be 100 and the set area can be 400 square millimeters.

[0092] S4, registering the preoperative femur and tibia model point cloud of the patient obtained in S1 and the postoperative femur and tibia model point cloud of the patient obtained in S3 after removing the points on the actual osteotomy plane;

[0093] Specifically, in the present invention, the point cloud of the patient's preoperative femur and tibia model obtained in S1 is defined as the preoperative point cloud, and the point cloud of the patient's postoperative femur and tibia model obtained in S3 after removing the points on the actual osteotomy plane is defined as the postoperative point cloud, which includes:

[0094] S41. Randomly select at least three point pairs at corresponding positions of the pre-operative point cloud and the post-operative point cloud, and calculate the initial transformation matrix of the two.

[0095] Since most of the preoperative bony landmarks are removed during surgery, it is necessary to select at least three point pairs at the corresponding positions of the patient's preoperative complete femur and tibia model point cloud and the postoperative incomplete femur and tibia model point cloud to calculate the initial transformation matrix of the two.

[0096] In the present invention, since the error of the initial point pair has a significant impact on the final registration result, the initial transformation matrix may cause the subsequent ICP algorithm to fall into a local optimal solution during the optimization process. Therefore, before the ICP algorithm performs fine registration, the initial transformation matrix can also be optimized as follows:

[0097] 1) According to the initial transformation matrix obtained above, the postoperative point cloud is transformed to the same reference as the preoperative point cloud;

[0098] 2) Randomly select three points in the preoperative point cloud whose distance between each pair is greater than a set value, and find the point in the transformed postoperative point cloud that is closest to the selected point, thereby obtaining three sets of point pairs;

[0099] In the present invention, the set value can be 50 mm.

[0100] 3) Repeat step (2) for a set number of times, and perform registration after selecting three sets of point pairs each time, calculate the corresponding transformation matrix, and apply it to the postoperative point cloud;

[0101] In the present invention, the set number of times can be 50 times.

[0102] 4) Calculate the average distance between the three points selected in the preoperative point cloud and the corresponding points in the postoperative point cloud transformed in step (3), and select the transformation matrix corresponding to the minimum average distance as the initial transformation matrix between the final preoperative point cloud and the postoperative point cloud.

[0103] In the present invention, in order to achieve faster convergence of the ICP algorithm, voxel down sampling can be performed on the point cloud of the patient's postoperative femur and tibia model after removing the points on the actual osteotomy plane. Specifically, voxel down sampling can be used for voxel down sampling, and the initial transformation matrix is ​​optimized based on the point cloud of the patient's postoperative femur and tibia model after removing the points on the actual osteotomy plane after voxel down sampling.

[0104] In this embodiment, after voxel downsampling and removal of points on the actual osteotomy plane, the voxel block size in the point cloud of the patient's postoperative femur and tibia model is 35 mm×35 mm×35 mm.

[0105] S42. Using the initial transformation matrix obtained in S41 as an initial value, use the ICP algorithm to calculate the transformation matrix between the patient's preoperative femoral and tibia model point cloud and the patient's postoperative femoral and tibia model point cloud after removing the points on the actual osteotomy plane.

[0106] S5. Based on the registration in S4, the actual osteotomy plane obtained in S3 and the planned osteotomy plane obtained in S2 are transformed to the same reference, and the angle error and distance error between the actual osteotomy plane and the planned osteotomy plane are calculated, as well as the osteotomy error between the actual osteotomy amount obtained in S3 and the planned osteotomy amount obtained in S2;

[0107] In the present invention, the planned osteotomy plane and the actual osteotomy plane may both contain multiple planes, and the normal angles between each plane are quite different. However, the normal angles between the corresponding planes in the planned osteotomy plane and the actual osteotomy plane are relatively small. Therefore, the angular error between the planned osteotomy plane and the actual osteotomy plane can be determined by calculating the normal angles between the corresponding planes in the planned osteotomy plane and the actual osteotomy plane.

[0108] In a specific embodiment of the present invention, during knee replacement surgery, the distal femoral osteotomy plane FB, the posterior femoral condyle osteotomy plane FR, and the tibial plateau osteotomy plane TP obtained by osteotomy of the femur directly determine the placement of the prosthesis. Figure 2 As shown, each plane has a corresponding amount of osteotomy, which in turn determines the patient's postoperative outcome. Therefore, this embodiment verifies the osteotomy accuracy by calculating the osteotomy plane error and osteotomy amount error of these three planes.

[0109] Specifically, the steps include:

[0110] S51, based on the registration in S4, transforming the actual osteotomy plane in the postoperative femoral and tibia model point cloud of the patient identified in S3, and then transforming the actual osteotomy plane to the same reference as the preoperative femoral and tibia model point cloud of the patient, and finding a plane corresponding to the distal femoral osteotomy plane FB, the posterior femoral condyle osteotomy plane FR, and the tibial plateau osteotomy plane TP in the planned osteotomy plane;

[0111] In the present invention, since the entire surgical system is controllable and the normal differences between the three planes of the distal femoral osteotomy plane FB, the posterior femoral condyle osteotomy plane FR and the tibial plateau osteotomy plane TP are relatively large, the plane in the actual osteotomy plane that is closest to the distal femoral osteotomy plane FB, the posterior femoral condyle osteotomy plane FR and the tibial plateau osteotomy plane TP in the planned osteotomy plane and has the smallest normal angle can be selected as the corresponding plane, and then the planes corresponding to the distal femoral osteotomy plane FB, the posterior femoral condyle osteotomy plane FR and the tibial plateau osteotomy plane TP in the planned osteotomy plane can be selected in the actual osteotomy plane respectively.

[0112] S52, calculating the angle error and distance error between the actual osteotomy plane and the planned osteotomy plane based on the planes in the actual osteotomy plane obtained in S51 that correspond to the distal femoral osteotomy plane FB, the posterior femoral condyle osteotomy plane FR, and the tibial plateau osteotomy plane TP in the planned osteotomy plane;

[0113] In the present invention, the angle error and distance error between the calculated actual osteotomy plane and the planned osteotomy plane include:

[0114] 1) Calculate the average distance between all intersection points of the distal femoral osteotomy plane FB in the planned osteotomy plane and the patient's postoperative femoral and tibia model point cloud and the corresponding plane in the actual osteotomy plane, and calculate the angle between the normal direction of the distal femoral osteotomy plane FB in the planned osteotomy plane and the normal direction of the corresponding plane in the actual osteotomy plane;

[0115] 2) Calculate the average distance between all intersection points of the femoral posterior condyle osteotomy plane FR in the planned osteotomy plane and the patient's postoperative femoral and tibial model point cloud and the corresponding plane in the actual osteotomy plane, and calculate the angle between the normal direction of the femoral posterior condyle osteotomy plane FR in the planned osteotomy plane and the normal direction of the corresponding plane in the actual osteotomy plane;

[0116] 3) Calculate the average distance between all intersection points of the tibial plateau osteotomy plane TP in the planned osteotomy plane and the patient's postoperative femoral and tibial model point cloud and the corresponding plane in the actual osteotomy plane, and calculate the angle between the normal direction of the tibial plateau osteotomy plane TP in the planned osteotomy plane and the normal direction of the corresponding plane in the actual osteotomy plane;

[0117] The above three sets of distances and angles together describe the angle error and distance error between the actual osteotomy plane and the planned osteotomy plane.

[0118] S53, combining the actual osteotomy amount of the plane corresponding to the distal femoral osteotomy plane FB, the posterior femoral condyle osteotomy plane FR, and the tibial plateau osteotomy plane TP in the planned osteotomy plane in the actual osteotomy plane obtained by S3 and S51, and combining it with the planned osteotomy amount obtained by S2 to calculate the osteotomy amount error;

[0119] The calculation of the amount of osteotomy specifically includes:

[0120] 1) Calculate the distances from the distal femoral medial condyle point and the distal femoral lateral condyle point to the distal femoral osteotomy plane FB in the planned osteotomy plane to obtain the distal femoral medial condyle osteotomy amount and the distal femoral lateral condyle osteotomy amount;

[0121] 2) Calculate the distances from the medial and lateral points of the posterior femoral condyle to the posterior femoral condyle osteotomy plane FR in the planned osteotomy plane to obtain the medial and lateral osteotomy amounts of the posterior femoral condyle;

[0122] 3) Calculate the distances from the medial and lateral concave points of the tibial plateau to the tibial plateau osteotomy plane TP in the planned osteotomy plane to obtain the medial and lateral tibial plateau osteotomy amounts.

[0123] Therefore, the corresponding osteotomy amount error can be calculated based on the corresponding planned osteotomy amount obtained in S3.

[0124] The present invention distinguishes between surgical plans with imaging and surgical plans without imaging, and adopts different methods to obtain the planned osteotomy plane. After obtaining the actual osteotomy plane in the postoperative image, the postoperative image without the actual osteotomy plane is aligned, thereby achieving alignment with the point cloud of the defective bone model obtained after the operation, ensuring the accuracy of the alignment. For both surgical plans with imaging and those without imaging, the present invention can timely evaluate the surgical effect of knee replacement surgery, and can objectively judge the success of the surgery. It has wide applicability in clinical practice.

[0125] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention (including the claims) is limited to these examples. Within the scope of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0126] The embodiments of the present invention are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for verifying osteotomy accuracy in knee replacement surgery, characterized in that: include: S1. Obtain the patient's preoperative femoral and tibia model point cloud and mark the anatomical bony landmarks in it; S2, determine whether it is a non-imaging surgical plan; If yes, obtain the patient's intraoperative femoral and tibia model point clouds and the anatomical bony landmarks therein, and use them to obtain the planned osteotomy plane and planned osteotomy amount. At the same time, align the patient's preoperative femoral and tibia model point clouds with the intraoperative femoral and tibia model point clouds; Otherwise, the planned osteotomy plane and planned osteotomy amount are obtained according to S1; S3, obtaining the postoperative femoral and tibia model point cloud of the patient after osteotomy, identifying the actual osteotomy plane therein, and then calculating the actual osteotomy amount, while removing the points on the identified actual osteotomy plane; S4, registering the patient's preoperative femoral and tibia model point clouds with the postoperative femoral and tibia model point clouds after removing points on the actual osteotomy plane; S5. Based on the registration in S4, the actual osteotomy plane obtained in S3 and the planned osteotomy plane obtained in S2 are transformed to the same reference, and the osteotomy plane error and the osteotomy amount error are calculated accordingly.

2. The osteotomy accuracy verification method for knee replacement surgery according to claim 1, characterized in that: In S3, the actual osteotomy plane is identified by acquiring normal information and topological structure of the points in the point cloud of the patient's postoperative femur and tibia model.

3. The osteotomy accuracy verification method for knee replacement surgery according to claim 2, characterized in that: In S3, identifying the actual osteotomy plane specifically includes: S31, using a voxel block of a set size to perform voxel downsampling on each point in the postoperative femoral and tibia model point cloud of the patient, traversing each sampling point obtained after the voxel downsampling, and classifying a sampling point and points within a set range thereof as the same region of interest; S32, clustering the region of interest obtained in S31 to obtain several candidate regions, and fitting several candidate planes based on the regions; S33, screening a set of planes that meet the requirements from the candidate planes obtained in S32 as the final actual osteotomy plane.

4. The osteotomy accuracy verification method for knee replacement surgery according to claim 3, characterized in that: In the S32, clustering the regions of interest includes: (1) Obtain the points and their normal vectors in each region of interest, fit each region of interest to a plane, and retain the regions of interest whose corresponding planes meet the set conditions; (2) Clustering the regions of interest retained in step (1), merging them to obtain several candidate regions, and fitting each candidate region to obtain several candidate planes.

5. The osteotomy accuracy verification method for knee replacement surgery according to claim 4, characterized in that: In the step (1), the points and normal vectors in each region of interest are obtained, and each region of interest is fitted into a plane using the least squares method by calculating the sum of the normal vectors of all points.

6. The osteotomy accuracy verification method for knee replacement surgery according to claim 4, characterized in that: In step (1), the setting conditions are specifically: the distance between all points in each region of interest and the corresponding plane is less than a first set distance, and the angle between the sum of the normal vectors of all points in each region of interest and the normal vector of the corresponding plane is less than a first set angle.

7. The osteotomy accuracy verification method for knee replacement surgery according to claim 4, characterized in that: In step (2), the clustering conditions include: For any two regions of interest A and B, the distance between the points in A and the points in B is traversed and calculated. When the distance between the nearest points in A and B, that is, the aforementioned minimum distance, is less than the second set distance, and the angle between the normal vectors of the planes corresponding to A and B is less than the second set angle, and at the same time, the average distance between all points in A and the plane corresponding to B is less than the third set distance, then the regions of interest A and B are considered to be the same category, and the regions of interest A and B are merged into a new region as a candidate region.

8. The osteotomy accuracy verification method for knee replacement surgery according to claim 3, characterized in that: In S33 , the screening requirement is that the number of points corresponding to the candidate plane is greater than a set number, and the area of ​​the minimum circumscribed rectangle of all points corresponding to the candidate plane is greater than a set area.

9. The osteotomy accuracy verification method for knee replacement surgery according to claim 1, characterized in that: In said S4, the point clouds of the patient's preoperative femur and tibia models are defined as preoperative point clouds, and the point clouds of the patient's postoperative femur and tibia models after removing the points on the actual osteotomy plane are defined as postoperative point clouds; Then the S4 includes: S41. Randomly select at least three point pairs at corresponding positions of the pre-operative point cloud and the post-operative point cloud, and calculate the initial transformation matrix of the two. S42. Using the initial transformation matrix obtained in S41 as an initial value, use the ICP algorithm to calculate the transformation matrix between the pre-operative point cloud and the post-operative point cloud.

10. The osteotomy accuracy verification method for knee replacement surgery according to claim 9, characterized in that: In S41, after obtaining the initial transformation matrix, the step of optimizing the initial transformation matrix is ​​also included, including: 1) transforming the postoperative point cloud according to the initial transformation matrix to the same reference as the preoperative point cloud; 2) Randomly select three points in the preoperative point cloud whose distance between each pair is greater than a set value, and find the point in the transformed postoperative point cloud that is closest to the selected point, thereby obtaining three sets of point pairs; 3) Repeat step 2) a set number of times, selecting three point pairs each time for registration, calculating the corresponding transformation matrix, and applying it to the postoperative point cloud; 4) Calculate the average distance between the three points selected in the preoperative point cloud and the corresponding points in the postoperative point cloud transformed in step 3), and select the transformation matrix corresponding to the minimum average distance as the initial transformation matrix between the final preoperative point cloud and the postoperative point cloud.

11. The osteotomy accuracy verification method for knee replacement surgery according to claim 1, characterized in that: In the above-mentioned S5, the osteotomy accuracy is verified by calculating the osteotomy plane error and the osteotomy amount error of the distal femoral osteotomy plane, the posterior femoral condyle osteotomy plane, and the tibial plateau osteotomy plane.

12. The osteotomy accuracy verification method for knee replacement surgery according to claim 11, characterized in that: The calculation of the osteotomy plane error and the osteotomy amount error of the distal femoral osteotomy plane, the posterior femoral condyle osteotomy plane, and the tibial plateau osteotomy plane includes: S51. Based on the registration in S4, transform the actual osteotomy planes in the point clouds of the postoperative femoral and tibia models of the patient, transform the actual osteotomy planes to the same reference as the planned osteotomy planes, and search for planes corresponding to the distal femoral osteotomy plane, the posterior femoral condyle osteotomy plane, and the tibial plateau osteotomy plane in the planned osteotomy planes. S52, calculating the angle error and distance error between the actual osteotomy plane and the planned osteotomy plane according to S51; S53: Combine S3 and S51 to obtain the actual osteotomy amount of the corresponding plane, and combine it with the planned osteotomy amount obtained in S2 to calculate the osteotomy amount error.

13. The osteotomy accuracy verification method for knee replacement surgery according to claim 12, characterized in that: In the step S52, the angle error and distance error between the actual osteotomy plane and the planned osteotomy plane are calculated as follows: Calculate the average distance between all intersection points of the distal femoral osteotomy plane in the planned osteotomy plane and the patient's postoperative femoral and tibia model point cloud and the corresponding plane in the actual osteotomy plane, and calculate the angle between the normal direction of the distal femoral osteotomy plane in the planned osteotomy plane and the normal direction of the corresponding plane in the actual osteotomy plane; Calculate the average distance between all intersection points of the femoral posterior condyle osteotomy plane in the planned osteotomy plane and the patient's postoperative femoral and tibial model point cloud and the corresponding plane in the actual osteotomy plane, and calculate the angle between the normal direction of the femoral posterior condyle osteotomy plane in the planned osteotomy plane and the normal direction of the corresponding plane in the actual osteotomy plane; Calculate the average distance between all intersection points of the tibial plateau osteotomy plane in the planned osteotomy plane and the patient's postoperative femoral and tibia model point cloud and the corresponding plane in the actual osteotomy plane, and calculate the angle between the normal direction of the tibial plateau osteotomy plane in the planned osteotomy plane and the normal direction of the corresponding plane in the actual osteotomy plane; The above three sets of distances and angles together describe the angle error and distance error between the actual osteotomy plane and the planned osteotomy plane.

14. The osteotomy accuracy verification method for knee replacement surgery according to claim 12, characterized in that: In S53, the actual amount of osteotomy in the corresponding plane is obtained by combining S3 and S51, specifically: Calculate the distances from the distal femoral medial condyle point and the distal femoral lateral condyle point to the distal femoral osteotomy plane in the planned osteotomy plane to obtain the distal femoral medial condyle osteotomy amount and the distal femoral lateral condyle osteotomy amount; Calculate the distances from the medial point of the posterior femoral condyle and the lateral point of the posterior femoral condyle to the posterior femoral condyle osteotomy plane in the planned osteotomy plane to obtain the medial and lateral osteotomy amounts of the posterior femoral condyle; The distances from the medial concave point of the tibial plateau and the lateral concave point of the tibial plateau to the tibial plateau osteotomy plane in the planned osteotomy plane were calculated to obtain the medial tibial plateau osteotomy amount and the lateral tibial plateau osteotomy amount.

15. The osteotomy accuracy verification method for knee replacement surgery according to claim 1, characterized in that: In S2, the point cloud of the patient's preoperative femur and tibia model is registered with the point cloud of the femur and tibia model during the operation, as follows: The intraoperative anatomical bony landmarks and the preoperative anatomical bony landmarks obtained by S1 form corresponding point pairs, and the initial transformation matrix between the patient's preoperative femur and tibia model point cloud and the intraoperative femur and tibia model point cloud is calculated. Based on this, the ICP algorithm is used to align the patient's preoperative femur and tibia model point cloud with the intraoperative femur and tibia model point cloud to obtain the transformation matrix between the two.

Images